Stock Price Prediction using LSTM

Abstract

The movement of stock prices is non-linear and complicated. In this study, we compared and analyzed various neural network forecasting methods based on real problems related to stock price demand forecasting. We ultimately selected the LSTM (Long Short-Term Memory) [1] neural network as traditional RNN’s long-term reliance is improved by LSTM, which substantially enhances prediction accuracy and stability. The practicality of this method and the pertinence of the model are then inspected, and final conclusions are drawn through a detailed examination of stock price forecasts using LSTM neural networks optimized by RNN algorithms. Past information has proven to be extremely predominant to investors as the basis for financing resolution. Previous studies have used open and close prices as vital predictors of financial markets, but utmost highs and lows may provide extra information regarding future price actions. Hence, two representative stock indexes of the Indian stock market were selected for the survey, and the main data collected from them are open, closed, lowest, highest, date, and everyday transaction size. The outcome shows that the LSTM model has few restraints, including a forecast time lag, but you can use the attention level to foretell stock prices. Its primary idea is to analyze stock market historical information and find the role of time series by digging deeper into its central rules.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 29 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.